Update repo to isort v5 (#6686)

* Run new isort

* More changes

* Update CI, CONTRIBUTING and benchmarks

examples/seq2seq/bertabs/run_summarization.py

@@ -44,9 +44,10 @@ def evaluate(args):
         reference_summaries = []
         generated_summaries = []
 
-        import rouge
         import nltk
 
+        import rouge
+
         nltk.download("punkt")
         rouge_evaluator = rouge.Rouge(
             metrics=["rouge-n", "rouge-l"],